LWWJ303-14 January 26, 2007 1:23 Char Count= 0

Advances in Nursing Science

Vol. 30, No. 1, pp. 52–60
Copyright  c 2007 Wolters Kluwer Health | Lippincott Williams & Wilkins

A Trojan Horse for Positivism?

A Critique of Mixed Methods
Research
Lynne S. Giddings, PhD, RGON, MN; Barbara M. Grant, PhD

Mixed methods research is captured by a pragmatically inflected form of postpositivism. Al-

though it passes for an alternative methodological movement that purports to breach the
divide between qualitative and quantitative research, most mixed methods studies favor the
forms of analysis and truth finding associated with positivism. We anticipate a move away
from exploring more philosophical questions or undertaking modes of enquiry that challenge
the status quo. At the same time, we recognize that mixed methods research offers particular
strengths and that, although it serves as a Trojan Horse for positivism, it may productively carry
other paradigmatic passengers. Key words: mixed methods, qualitative research, quanti-
tative research, paradigms, postactivism, pragmatism

M IXED methods research is being pro-

moted as a practice that breaches di-
vide between qualitative and quantitative re-
search. Here, we join the current debate in
the nursing literature1–5 by advancing the
view that mixed methods is a Trojan Horse
for positivist enquiry, depending for its appeal
on a pragmatic orientation. The critique of
mixed methods research offered here arises
from our shared concern at its contempo-
rary positioning as the “third methodological
movement.”6(pix) Such a positioning is fraught
with theoretical and political complexities:
From the School of Nursing, Faculty of Health &
Environmental Sciences, Auckland University of
Technology (Dr Giddings); and the Centre for
Professional Development, University of Auckland
(Dr Grant), Auckland, New Zealand.
We are grateful for feedback from the Young and Rest-
less Scholars Writing Group (The University of Auck-
land), Claire-Louise McCurdy for passing her editorial
eye over the manuscript, and the academic women
who attended the Tauhara Writing Retreats (Taupo,
Aotearoa, New Zealand) between 2003 and 2005 who
patiently listened to and commented on our developing
ideas and musings on mixed methods research.
Corresponding author: Lynne S. Giddings, PhD, RGON,
MN, School of Nursing, Faculty of Health & Environ-
mental Sciences, Auckland University of Technology,
Auckland 1142, New Zealand (e-mail: Lynne.giddings@
aut.ac.nz).
52
in particular, it shores up the argument that
research is value neutral rather than grap-
pling with its painful politics. In what fol-
lows, and drawing on a comprehensive re-
view of recent nursing research (L.S.G. and
L. A. Williams, unpublished data, 2006), we
argue that mixed methods research in nurs-
ing and health generally, and increasingly in
the social sciences including education, has
been captured by a pragmatic postpositivism
and that such capture secures mixed meth-
ods within the broader positivist project to
know the world in particular ways. The effect
of this capture is to reinstall the marginaliza-
tion of other forms of knowing. Moreover, the
resultant narrowing of focus means more cir-
cumscribed fields of values at play, questions
being asked, forms of data being collected,
modes of analysis being undertaken, and pos-
sible outcomes being generated.
The trend toward reinstalling positivist re-
search as the methodology “of choice” has
been noted by others. For example, Patti
Lather notes “a resurgent positivism and gov-
ernmental imposition of experimental design
as the gold standard in [educational] research
methods”7(p35) in North America, as does
Robert Donmoyer8 who notes that this has
been a recent and rather dramatic change.
In her critique of evidence-based practice,
LWWJ303-14 January 26, 2007 1:23 Char Count= 0
A Critique of Mixed Methods Research
Sue Clegg points toward similar trends in at-
titudes toward social policy research in the
United Kingdom as they increasingly become
subject to the logic of “what works.”9(p416) A
cursory survey of Web sites for health fund-
ing bodies by one of the authors (L.S.G.)
shows the prevalence of terms such as “holis-
tic,” “integrated,” “multi-disciplinary,” “inter-
disciplinary,” “cross-disciplinary,” and “collab-
orative,”all code words signaling a preference
for mixed methods research designs.
In what follows, we explain our posi-
tion by exploring how mixed methods re-
search has been captured by pragmatic post-
positivism. In particular, we tease out the
theoretical continuities between positivism
and postpositivism to clarify our claim that
the latter remains within the paradigmatic
purview of the former. In previous work, we
have laid out a schematic view of researcher
paradigms in the health and social sciences.10
The critique presented here draws on the
schema that comprised 4 paradigms: posi-
tivist, interpretivist, radical (also called crit-
ical), and poststructural. In that work, we
also acknowledged the emergence of indige-
nous research—for example, Kaupapa Maori
research in Aotearoa, New Zealand—as a pos-
sible further paradigm. We go on to attend
to two related conceptual confusions that are
critical to this debate. Finally, we acknowl-
edge the strengths of mixed methods research
and explore the ways it can serve paradigms
other than that most closely associated with
science.
THE PRAGMATIC AND POSTPOSITIVIST
CAPTURE OF MIXED METHODS
RESEARCH
Pragmatism and postpositivism have been
significant influences on the modern mixed
methods movement. Although both have
been viewed as paradigms or “knowledge
claims,”11(pp4–12) we see pragmatism as an
ideological position available within any
paradigm rather than a paradigm in its own
right. In practice, proponents of mixed meth-
53
ods research are often pragmatists in orienta-
tion. Rather than focusing on epistemological
integrity, they emphasize the importance of
getting “the job done.”12(p101) It is the prac-
ticality of the designs and their “wide range
of uses”13(p364) that is valued. For these rea-
sons, pragmatism’s conjuncture with postpos-
itivism is particularly fitting: as modes of re-
search enquiry, both are marked by a lack of
theoretical reflexivity and both value eclec-
ticism in choice of methods, although for
different reasons.
In contrast to pragmatism, postpositivism
is of the order of a paradigm: it is a distinc-
tive development within the paradigm of pos-
itivism, arising from the recognition of pos-
itivism’s ideological and practical limitations
for some forms of research, including nurs-
ing, health, and social science research. Cru-
cially, postpositivist thinking also contributed
to the emergence of the alternative research
paradigms identified above, making space for
standpoints that no longer believed in the
possibility of understanding life from an ob-
jective point of view. Our critique is not di-
rected at postpositivism per se but rather the
way in which it works stealthily to entrench
positivism.
In the short history of health and social
science research, the views of scientific pos-
itivism have dominated, although not with-
out criticism. For instance, Wilhelm Dilthey
(1833–1911) and Edmund Husserl (1859–
1938), whose ideas became the basis for mod-
ern phenomenology, argued that the scien-
tific method was inappropriate for studying
human phenomena: “What [life] is cannot be
expressed in a simple formula or explanation.
Thought cannot fully go behind life, for it is
the expression of life.”14(p25)
In the second half of the 20th century,
there were multiple challenges to positivist
science from the protagonists of other com-
peting paradigms such as feminists, critical
social theorists, and poststructuralists. But
probably most influential were the even
earlier criticisms in the 1960s and 1970s
within the positivist paradigm itself from a
group of philosophers of science, including
LWWJ303-14 January 26, 2007 1:23 Char Count= 0

54 ADVANCES IN NURSING SCIENCE/JANUARY–MARCH 2007

Karl Popper,15 Thomas Kuhn,16 Stephen
Toulmin,17 and Paul Feyerabend.18 Although
these thinkers came from different per-
spectives, collectively they destabilized the
positivist notions of absolute truth, prov-
able hypotheses, and unbiased, value-free re-
searchers. Their criticisms meant that posi-
tivist science began to lose the high ground.
Indeed, it was these postpositivist thinkers
that made the ideological space for the
emergence of qualitative methodologies in
the 1980s, and, in turn, the modern-day
appearance of mixed methods research.
Postpositivism—the “post” signaling a devel-
opment of positivism—emerged as a more
moderate form of positivism, but one that is
neither accepted nor understood by all posi-
tivists. Indeed, it is the case that the many pos-
itivist researchers do not understand them-
selves in an epistemological or an ontological
sense, a phenomenon not unusual in the self-
awareness of dominant social groups.
In many ways, the underpinning assump-
tions of postpositivism are continuous with
positivism, as Table 1 makes plain. Before
we go any further, a caution: by catego-
rizing research activities into a neat list of
philosophical assumptions and characteris-
tics, we are at risk of stereotyping the so-
labeled researchers. We are aware of the di-
versity of opinion among postpositivists, as
indeed postpositivist colleagues who critique
our work continually remind us! Our catego-
rizing framework is offered to show the posi-
tivist antecedents of postpositivism in order
to support our argument that the latter re-
mains within the general worldview of the for-
mer.
Table 1 summarizes the continuities and
discontinuities between the key philosophi-
cal assumptions of positivism and those of
postpositivism. A core fundamental positivist
assumption is that of determinism, the be-
lief that effects have a determinable cause and

Table 1. Comparing the philosophical underpinnings of positivism and postpositivism

Philosophical
assumptions Positivist position Postpositivist position

Determinism Effects have determinable causes and
actions have predictable outcomes
Reductionism Experience can be reduced to a
discrete set of ideas or concepts
that can be described and tested
Objectivism Reality exists out in the world and
can be observed, measured, and
understood. Objectivity is the goal
of the researcher
Theory verification Theory is universal and generalizable
through a process of proving
hypotheses
Role of evidence Evidence is required to establish
“truth”
Scientific method The scientific method is the best way
of knowing about the world and
seeking evidence to solve
problems
Effects and outcomes are the result of
a complex array of interactive
causative and outcome factors
Experience can be described
conceptually and tested, but the
unpredictable and contradictory
nature of human experience needs
to be factored in
Reality is socially and culturally
constructed and can be observed
and measured. Researcher
objectivity is impossible
Theory remains open to verification
through a process of supporting
hypotheses
Evidence establishes degrees of
probability that something is “true”
There is not one method: choice of
method (qualitative and
quantitative) is guided by the
research question
LWWJ303-14 January 26, 2007 1:23 Char Count= 0
A Critique of Mixed Methods Research
actions have predictable outcomes. Postposi-
tivists maintain this assumption in a modified
form: rather than assuming a linear process
of cause and effect, they perceive outcomes
as the result of a complex array of causative
factors that are in interaction with their out-
comes. Postpositivists also maintain the pos-
itivist assumption of reductionism, the belief
that experience can be reduced to a discrete
set of ideas or concepts that can be described
and tested. Again, however, this assumption is
modified: postpositivists factor in the unpre-
dictable and contradictory nature of human
experience. Another key positivist assump-
tion, objectivism, is the belief that reality ex-
ists out in the world and can be observed,
measured, and understood. On this point,
however, postpositivists diverge from posi-
tivists significantly: they tend to argue that
reality is socially and culturally constructed
and researcher objectivity is impossible. Dif-
ferent postpositivist researchers, though, take
a range of stances on this issue: at the most
positivist-inflected end of the spectrum, they
argue that the researcher must strive to be as
neutral as possible, while at the other end,
they argue that the researcher cannot be neu-
tral because she is in relationship with what
she is researching.
Postpositivist divergence from “pure” pos-
itivism is found in yet other key assump-
tions: postpositivists maintain the assump-
tion of theory verification (the belief that
laws and theories can explain various reali-
ties but that they need to remain open to
verification to establish “truth”), but tend to
talk about “supporting” rather than “proving”
hypotheses. The role of evidence is to es-
tablish a high degree of probability, rather
than certainty, that something is “true,”hence
the term “probabilistic evidence.”19(p14) This
represents a critical shift in thinking away
from the positivist assumption that theory can
be universal and generalizable. Postpositivists
also maintain the positivist assumption that
the scientific method is best; however, they
believe that choice of method is guided by
the research question and that research can
55
incorporate multiple methods, including non-
traditional ones, especially for triangulation.
Tashakkori and Teddlie define triangulation
as the “combination and comparisons of mul-
tiple data sources, data collection and analysis
procedures, research methods, and/or infer-
ences that occur at the end of a study.”6(p717)
In other words, triangulation offers ways to
verify and confirm findings so the researcher
has confidence and some certainty in the
conclusions made. Although methods are still
the processes by which “truth” can be estab-
lished, they are always open to challenge. The
research process needs to show evidence of
control for bias, with reliability and validity
standards used to ensure rigor. This repre-
sents another critical shift in thinking away
from the positivist assumption that the tradi-
tional scientific method is the way to establish
truth: indeed, the belief that qualitative meth-
ods have something to add to the findings of
quantitative ones underpins the postpositivist
uptake of mixed methods research.
In spite of these challenges to some of pos-
itivism’s basic tenets, postpositivism is funda-
mentally an extension of rather than a break
from the positivist paradigm. Scientific meth-
ods and principles are still accepted as the
best ways for discovering true knowledge and
solving problems. However, like their posi-
tivist kin, postpositivist researchers rarely ac-
knowledge the philosophical and theoretical
underpinnings of their research. A recent
survey of around 140 mixed methods re-
search articles featuring in nursing journals
between 1998 and July 2005 showed the ma-
jority of them take an implicit rather than ex-
plicit postpositivist orientation toward mixed
methods.20 We have already remarked that
such an omission can be understood as char-
acteristic of a dominant culture that does not
need to explain itself: the absence of reflex-
ivity about deeply held assumptions further
serves to maintain the dominance of the posi-
tivist worldview in the health and social sci-
ences. It is in this sense that mixed meth-
ods can be understood as a Trojan Horse for
positivism.
LWWJ303-14 January 26, 2007 1:23 Char Count= 0
56 ADVANCES IN NURSING SCIENCE/JANUARY–MARCH 2007
WHAT IS ‘‘MIXED’’ IN MIXED METHODS
RESEARCH?
The lack of explicitness about paradigmatic
positioning extends to other aspects of mixed
methods practice—in particular, there is of-
ten a lack of clear understanding of just what
is mixed—methods or methodologies. Our at-
tention now turns to this arena of theoretical
and political confusion.
The ingredient that is most commonly
mixed in mixed methods research is the meth-
ods, not the methodologies, and the methods
are mixed in the quite specific sense that both
qualitative and quantitative ones are used.
Because of this, some proponents of mixed
methods research argue that this approach to
research represents the “best of both worlds,”
usually understood as the two worldviews of
positivism/quantitative research and interpre-
tivism/qualitative research. This claim arises
from two critical and persistent confusions:
the first is a misunderstanding over the differ-
ence between the ideas of methodology and
method and the second is that over the status
of the terms qualitative and quantitative.
First, the methodology/method pair. The
more abstract of the two terms, methodol-
ogy, refers to the theoretical assumptions and
values that underpin a particular research ap-
proach. For this reason, methodologies be-
long within certain paradigms, although
they can be adapted to work in others—for
example, ethnography was originally an in-
terpretivist methodology but has since been
adapted for use within the critical/radical
paradigm as “critical ethnography.”Moreover,
particular methodologies are often associated
with specific disciplines—for example, al-
though ethnography is characteristic of so-
cial anthropology, nursing has subsequently
developed and applied it within certain prac-
tice contexts. Likewise, nursing research has
taken up the methodologies of phenomenol-
ogy and grounded theory that originally devel-
oped within other disciplines. Distinctively,
methodology is a thinking tool that guides
how a researcher frames her research ques-
tion and how she decides on what methods
and forms of data analysis to use.
Methods, in contrast, are much more con-
crete and practical—they are the doing tools
for collecting and analyzing data. To illus-
trate, one commonly used method in ethnog-
raphy is “participant observation,” whereby
the researcher gathers information by spend-
ing time in the community under study. In
contrast, a phenomenologist is more likely to
use in-depth “conversations”with a small sam-
ple (6–8) of participants to explore the mean-
ings of a particular phenomenon. As tools,
methods are almost always a-paradigmatic,
and therefore any given method may be used
in the service of any paradigm. In practice,
however, some methods are closely iden-
tified with particular methodologies within
certain paradigms. For example, the open-
ended interview method is closely associated
with many of the methodologies of the in-
terpretivist paradigm, and the survey method
with those of the positivist/postpositivist
paradigm.
Inability to distinguish between method-
ology and method can lead to confused
thinking and practice on the part of many
mixed methods researchers. This in turn
has given rise to charges of methodologi-
cal “slurring”21,22 and internal inconsistency
within research design. Moreover, sometimes
there are claims of methodological triangula-
tion in the literature22 —that is, researchers
who say they are bringing together the find-
ings of 2 or more methodologies in or-
der to strengthen their study and, indeed,
often to access funding opportunities. Al-
though mixed methodologies research does
offer the advantage of approaching a phe-
nomenon from different angles, more often
than not such research turns out to be mixing
the methods rather than the methodologies.
Where methodologies are mixed inside one
research project, there are often problems
with the commensurability of the findings.
This problem is particularly fraught when
those methodologies also cross paradigm
boundaries because then the underlying
LWWJ303-14 January 26, 2007 1:23 Char Count= 0
A Critique of Mixed Methods Research
assumptions and values are usually contra-
dictory. To give an example of this diffi-
culty: Jennifer Greene and Valerie Caracelli
assert the value of a “dialectic mixed meth-
ods” stance that intentionally includes differ-
ent paradigms in order to reach “better un-
derstanding” in research via an engagement
with the tensions that are invoked between
paradigms.12(p97) Yet, discussion of their illus-
trative case culminates in this way: “the dia-
logues in this evaluation were not actualised
as intended, and the reasons were largely re-
lated to values and politics.”12(p98) This un-
satisfactory outcome is what we would pre-
dict from mixed methodology research across
paradigms, because the significant differences
between paradigms are usually irresolvable,
depending as they do on deeply different as-
sumptions and values about the nature of the
world, the people within it, and the relations
between them.
Despite these difficulties, successful mixed
methodology research is possible under cer-
tain circumstances—for example, when both
methodologies lie within the same paradigm,
the underlying assumptions and values are
likely to be coherent and so the different
findings are less likely to be in tension with
one another. Methodological mixing across
paradigms can also be effective where one
methodology is in the service of another. In
such a case, and in contrast to the example
given above, one set of assumptions predom-
inates and the contradictions can be dealt
with openly. Feminist grounded theory, crit-
ical hermeneutics, and critical ethnography
are examples of such mixing; in all these
cases, methodologies from the interpretive
paradigm are framed by theoretical assump-
tions from the radical/critical paradigm, so
that the way the research is carried out from
the framing of the question to the presenta-
tion of the findings reflects the transformative
assumptions typical of the latter. Judith Wuest
and colleagues explain one version of this:
“When grounded theory and feminist theory
are used together, theoretical sensitivity is in-
fluenced by feminism: investigators are re-
sponsive to the ways that gender, culture,
57
class, ability, age, and sexual orientation are
revealed in the data and influence the varia-
tion in emerging theoretical concepts.”23(p258)
Finally, on this matter of methodology
and method, a related source of confusion
is that mixed methods research is some-
times referred to itself as a methodology.
In our view, to have the status of a method-
ology, mixed methods research design must
be described by a larger term that points to
the theoretical positioning underpinning it,
for example, descriptive (explanatory or ex-
ploratory) mixed methods, feminist partici-
patory mixed methods, ethnographic mixed
methods, and so on. In just such a vein,
the most recent version of Lather’s paradigm
chart lists “positivist mixed methods” and “in-
terpretive mixed methods”as distinctive prac-
tices within disparate paradigms.7(p37) Other-
wise, the term “mixed methods”simply refers
to a particular selection of methods and as
an unsituated practice, risks a lack of internal
consistency within the research design.
A second and critical confusion is the per-
ceived difference in status between the terms
“qualitative” and “quantitative.” These terms
are commonly used to describe the meth-
ods or methodologies of mixed methods re-
search. Historically, quantitative research has
been viewed as synonymous with positivism
and qualitative with interpretivism—hence
the association with methodology. More rad-
ically, some writers consider the terms to
refer to two research paradigms in and of
themselves.24 At the other extreme, they
are considered to be terms merely descrip-
tive of forms of data: quantitative data be-
ing numbers and statistics, and qualitative be-
ing words and narratives. However, like Egon
Guba and Yvonna Lincoln, we argue that
the two terms most usefully describe differ-
ent “types of methods”25(p105) that may be
used for data collection and analysis, and that
methods in this sense are “a-theoretical and
a-methodological.”26(p33) As a-theoretical (or
a-paradigmatic) doing tools, the methods of
research can be mixed without contradiction,
although, as we have remarked above, with-
out methodological awareness they may well
LWWJ303-14 January 26, 2007 1:23 Char Count= 0
58 ADVANCES IN NURSING SCIENCE/JANUARY–MARCH 2007
be used in ways that are at odds with each
other and one (usually quantitative) will then
come to dominate.
The likely dominance of quantitative data
is an outcome of the politics associated with
research paradigms. In this landscape, pos-
itivist science still holds the high ground.
For example, scrutinizing the available mixed
methods research literature, the reader may
gain the impression that qualitative research
is only exploratory to, or supportive of, quan-
titative research data. This intimates that qual-
itative research cannot stand on its own and
is only validated by being attached to a sci-
entific, quantitative, evidence-based method-
ology. This status difference is compounded
by another, deeper misunderstanding. Novice
and even well-established postpositivist re-
searchers may misconstrue the inclusion of a
qualitative method into their research design
as “doing qualitative research.”This reinforces
the classic “talking past each other” that oc-
curs so often between researchers from differ-
ent paradigms. For example, a postpositivist
researcher who has used a qualitative induc-
tive process to analyze her data may dismiss
the complexity and rigor of the work of a
Heideggerian hermeneutic phenomenologist
as unnecessarily complicating things.1 Such a
dismissal overlooks the distinctive underlying
assumptions that differentiate the paradigms
and guide their disparate methodologies. In
this way, the increasing popularity of mixed
methods research further muddies the waters
around the significant differences between
research paradigms and reasserts the domi-
nance of the positivist/postpositivist one.
GOING FORWARD WITH MIXED
METHODS RESEARCH
Messiness occurs when researchers do
not acknowledge their paradigmatic position-
ing. Moreover, the use of qualitative meth-
ods with a dusting of interpretive (qualita-
tive) concepts in a research report does not
make a positivist/scientific study inclusive of
the interpretive paradigm (the “best of both
worlds”). If paradigmatic assumptions are not
made explicit by a researcher, the ensuing
analysis may contain unprocessed contradic-
tions. In a review of 141 journal articles, Gid-
dings and Williams (unpublished data) found
that where there was a lack of goodness
of fit between findings, the qualitative ones
took the backseat in order to preserve the
“integrity” of the study’s conclusions. In ef-
fect, the qualitative findings are all too eas-
ily relegated to the position as “handmaiden”
of quantitative ones.7(p49) A contrasting re-
sponse would be to address the disparity be-
tween the findings as a way to refocus the
research question and/or the underlying re-
searcher assumptions and values concerning
the topic or construct.
Lather argues for the value of the “disjunc-
tive affirmation”7(p52) offered by paradigm
(and methodological) proliferation. This is a
way of thinking about paradigm incommensu-
rability and disagreement as neither cause for
war nor requiring reconciliation but as itself
a virtue: “Layering complexity, foreground-
ing problems, thinking outside easy intelli-
gibility and transparent understanding, the
goal is to move education [social] research
in many different directions in the hope that
more interesting and useful ways of knowing
will emerge.”7(p53) Further to this, Donmoyer
points out that paradigmatic incommensura-
bility is not the same as logical incompati-
bility and therefore a given researcher can
“conceivably employ different paradigms in
different circumstances and/or to accomplish
different goals.”8(p21) This is an argument for
the benefits of radical theoretical flexibility,
one that Donmoyer thinks is especially rele-
vant to public policy fields such as health, so-
cial policy, and education. In these fields, de-
cision makers must consider and balance a va-
riety of perspectives, some of which may be
antithetical, to address “a particular situation
or a particular point in time.”8(p23)
It follows from these arguments that there
may be research strengths to be gained by
combining qualitative and quantitative meth-
ods: these include a broader research fo-
cus and a wider variety of data collection
LWWJ303-14 January 26, 2007 1:23 Char Count= 0
A Critique of Mixed Methods Research
approaches that in turn enable richer descrip-
tions of a phenomenon to be gathered. Al-
though mixed methods research to date pri-
marily reflects postpositivist assumptions,1 it
does not have to be confined to this paradigm.
It is a research design that can be situ-
ated within others. For example, feminist re-
searchers have long shown creative flexibility
in their approaches to collecting and analyz-
ing data, able to utilize quantitative methods
in the service of radical/critical or poststruc-
turalist paradigms. Pamela Ironside’s study27
of implementing and evaluating narrative ped-
agogy using a sequential mixed methods
design is a case in point. Ironside used
a pretest-posttest questionnaire followed by
semistructured interviews to gather data from
her students in a research design where the
data from both methods are treated equally.
She first sets out the student responses to var-
ious items on the questionnaire, giving some
of the statistical results in support of conclu-
sions made, and then presents the qualitative
data thematically. In the analysis, Ironside at-
tempts to integrate the findings from the two
data sources, although the quantitative find-
ings were in a number of instances incongru-
ent and inconsistent with the qualitative ones.
Fruitfully, most of the discussion in the article
arises from this mismatch. Ironside’s work il-
lustrates the rich possibilities of mixed meth-
ods design where ultimately new questions
can be posed and new ways to explore them
are imagined.
CONCLUSION
As research in nursing, health, and social
sciences becomes more pragmatically driven
and unsituated mixed methods takes the fund-
ing high ground, we anticipate a move away
from exploring more philosophical questions
or undertaking modes of enquiry that chal-
REFERENCES
1. Giddings LS. Mixed-methods research: positivism
dressed in drag? J Res Nurs. 2006;11(3):195–203.
2. Johnstone PL. Mixed methods, mixed methodology
59
lenge the status quo. Yet, researchers need
to be aware of and speak to the epistemolo-
gical breaches that lie between positivism
and the paradigms that follow: effacing these
breaches forecloses creative possibilities for
working more appropriately and wisely with
the communities our research seeks to serve.
In spurring on such effacement, mixed meth-
ods research is a Trojan Horse for positivism,
reinstalling it as the most respected form of
social research, while at the same time—
through inclusion—neutralizing the oppo-
sitional potential of other paradigms and
methodologies that more commonly use qual-
itative methods. This tendency may have far-
reaching consequences for nursing and nurs-
ing research, influencing teaching curricula
(in particular, the scope and variety of post-
graduate research projects), faculty appoint-
ments, the kinds of research questions that
can be asked, possibilities for publication,
and, critically, funding decisions. Nursing re-
search has historically led the way in the de-
velopment of nonpositivist methodologies in
the health sciences—our warning is that this
position is threatened by the mixed methods
movement.
Mixed methods research does not need to
play this role. Situated mixed methods—that
is, a research practice conscious of its under-
lying assumptions, beliefs, and politics—may
be a powerful form of research inquiry, offer-
ing as it does the possibility of rich and con-
tradictory findings. Such contradictions may
in turn shift the ways in which we understand
key research constructs and practices as well
as remind us of the complexity of the social
world that we are attempting to understand
and intervene in. More subversively, the cur-
rently popularized rubric of “mixed methods”
may offer a Trojan Horse for other, more radi-
cal agendas to win funding from sources that
are not usually sympathetic to their cause—if
wily researchers play their cards right.
health services research in practice. Qual Health
Res. 2004;14(2):259–271.
3. Sandelowski M. Combining qualitative and
LWWJ303-14 January 26, 2007 1:23 Char Count= 0
60 ADVANCES IN NURSING SCIENCE/JANUARY–MARCH 2007
quantitative sampling, data collection, and anal-
ysis techniques in mixed-methods studies. Res Nurs
Health. 2000;9:279–287.
4. Williamson GR. Illustrating triangulation in mixed-
methods nursing research. Nurse Res. 2005;12(4):7–
18.
5. Gilbert T. Mixed methods and mixed methodologies:
the practical, the technical and the political. J Res
Nurs. 2006;11(3):205–217.
6. Tashakkori A, Teddlie C, eds. Handbook of Mixed
Methods in Social and Behavioral Research. Thou-
sand Oaks, Calif: Sage; 2003.
7. Lather P. Paradigm proliferation as a good thing to
think with: teaching research in education as a wild
profusion. Int J Qual Stud. 2006;19(1):35–57.
8. Donmoyer R. Take my paradigm . . . please! The
legacy of Kuhn’s construct in educational research.
Int J Qual Stud. 2006;19(1):11–34.
9. Clegg S. Evidence-based practice in educational re-
search: a critical realist critique of systematic review.
Br J Sociol Educ. 2005;26(3):415–428.
10. Grant BM, Giddings LS. Making sense of method-
ologies: a paradigm framework for the novice re-
searcher. Contemp Nurse. 2002;13:10–28.
11. Creswell JW. Research Design: Qualitative, Quanti-
tative, and Mixed Methods Approaches. 2nd ed. Lin-
coln, Neb: Sage; 2003.
12. Greene JC, Caracelli VJ. Making paradigmatic sense
of mixed methods practice. In: Tashakkori A, Ted-
dlie C, eds. Handbook of Mixed Methods in Social &
Behavioural Research. Thousand Oaks, Calif: Sage;
2003:91–110.
13. Morgan DL. Practical strategies for combining qual-
itative and quantitative methods: applications to
health research. Qual Health Res. 1998;8:362–367.
14. Polkinghorne D. Methodology for the Human Sci-
ences: Systems of Inquiry. Albany: State University
of New York Press; 1983.
15. Popper KR. The Logic of Scientific Discovery. New
York: Basic Books; 1959.
16. Kuhn T. The Structure of Scientific Revolutions.
Chicago: Chicago University Press; 1970.
17. Toulmin S. Forecasting and Understanding, Fore-
sight and Understanding: An Inquiry Into the Aims
of Science. Bloomingdale: Indiana University Press;
1961.
18. Feyerabend PK. Against Method: Outline of an An-
archistic Theory of Knowledge. London: Humanities
Press; 1975.
19. Polit DF, Beck CT, Hungler BP. Essentials of Nursing
Research: Methods, Appraisal, and Utilization. 5th
ed. Philadelphia: Lippincott; 2001.
20. Giddings LS, Williams LA. A challenge to the post
positivist domination of mixed methods research: a
review of nursing journals 1998–2005. Paper pre-
sented at: The Mixed Methods Conference 2006; July
8–11, 2006; Cambridge, UK.
21. Baker C, Wuest J, Stern PN. Method slurring: the
grounded theory/phenomenology example. J Adv
Nurs. 1992;17(11):1355–1360.
22. Magnusson C, Finnerty G, Pope R. Methodologi-
cal triangulation in midwifery education research.
Nurse Res. 2005;12(4):30–41.
23. Wuest J, Merritt-Gray M, Ford-Gilboe M. Regenerat-
ing family: strengthening the emotional health of
mothers and children in the context of intimate part-
ner violence. ANS Adv Nurs Sci. 2004;27(4):257–
274.
24. Blaxter L, Hughes C, Tight M. How to Research. 2nd
ed. Philadelphia: Open University Press; 2001.
25. Guba EG, Lincoln YS. Competing paradigms in
qualitative research. In: Denzin NK, Lincoln YS,
eds. Handbook of Qualitative Research. Thousand
Oaks, Calif: Sage; 1994:105–117.
26. Sarantakos S. Social Research. 3rd ed. New York: Pal-
grave Macmillan; 2005.
27. Ironside P. Trying something new: implementing and
evaluating narrative pedagogy using a multimethod
approach. Nurs Educ Perspect. 2003;24(30):122–
128.